As logged here by Kiwamu, there was a reported problem with the BS oplev. I've now gone through several hours of data and have found no instances of these small (on the order of 20 - 30 counts) changes in the QPD SUM causing the oplev pitch/yaw outputs to behave badly. After reading this alog, it sounds as if the BS oplev is behaving. In light of this I'm going to leave the oplev as is. I will continue to keep an eye on it, and if any problems are noticed during commissioning efforts please let me know.
Thanks, Jason. But it is still there. See the attached time series below:
These are time series data from this evening when the simple Michelson was locked on a dark fringe. The red arrows are to point the transient/glitch events.
As the oplev sum steps by about 10-20 counts, it kicks the BS through the oplev servos and causes a transient-type misaligment in the Michelson, which is visible in LSC-ASAIR_A_LF_OUT in the upper left of the plot. Since the oplev YAW and PIT signals are suppressed by the oplev damping loop, the transient does not always show up in these channels in an obvious way.
Also I am attaching a video of the AS camera from this evening when the simple Michelson was locked at a dark fringe. It is clear that the Michelson is suffering by this fast transient.